Remaining 214 patches from Jono

Sjoerd Simons sjoerd at luon.net
Sun Aug 24 21:44:06 BST 2014


Hey,

With 214 uploaded to experimental we should see some more testing of a
vaguely recent systemd. Some patches from Jon's branch i've left out as
i wasn't sure of them, would be great to get some feedback:
  http://anonscm.debian.org/cgit/pkg-systemd/systemd.git/log/?h=wip/sjoerd/experimental-jonno

Relevant patches are:

* e366cac Rework the semantics of restart job deadlock avoidance during early boot and shutdown.

The main patch here was sent to the systemd list, from the discussion it
seems upstream doesn't like that approach. I don't think this is an area
where we should diverge, so probably should be dropped 

* 507e02f Rework support for Debian-specific locale, keyboard and timezone conf files.

Splits up d/p/Add-back-support-for-Debian-specific-config-files.patch,
no apparent other changes. No opinion from my side really.

* 116e90e Order rcS.d init scripts before basic.target instead of sysinit.target.

Something to pull in. Moves the LSB scripts from rcS from sysinit.target
to basic.target. Didn't want to grab this for 214-1 as it changes the
status quo at the same time as moving to the sysv generator.

Needs checking whether there are no services currently in basic.target
that rely on bits from rcS

* cc46ad0 Add breaks on old syslog providers, they interact badly with the journal.

Adds breaks on syslog providers not supporting systemd, some
unversioned. Probably not something to grab (especially unversioned
breaks are harmful). But probably worth considering what the strategy is
for syslog providers that don't support journald for jessie?

* 1b5a4d8 Add breaks on old resolvconf, it interacts badly with systemd-resolved.

To drop, we don't enable systemd-resolved by default so i don't think
this add much.

* 1fc6118 Add breaks on old uuid-runtime, it lacks necessary systemd support.

To drop, this was a bug in the uuid package not something systemd broke specifically

* b1cbcd7 Rename one incredibly poorly named patch.

No opinion, but the original patch is indeed quite badly named.

* 1691d8f Fix various features that didn't work if /usr wasn't mounted in the initramfs.

Probably to drop, as the push is to always ensure /usr is mounted.
Although i don't think that has landed yet ?

* 68829ab Special-case some dependencies for sysv init scripts for better backwards compatibility.

Looks sensible to me, but could do with another review.

-- 
Sjoerd Simons <sjoerd at luon.net>



More information about the Pkg-systemd-maintainers mailing list